Researchers the Georgia Institute of Technology have devised a big data algorithm for customising the difficulty of video games.
The computational model predicts player's in-game performance, using big data to devise a level of difficulty they are capable of beating, teaching them new in-game skills as they do so. In theory, the model could be applied to training and educational content, allowing the big data model to provide benefits in the real world.
Further reading
Big Data Summit - Insight from the experts Runescape: A big data case study Rugby tries out data analytics
George Tech researchers developed a simple, turn-based game then used player scores to develop an algorithm to predict how others with a similar skill level would do, then adjusted the difficulty setting to cater for that.
"People come in playing video games with different skills, abilities, interests and even desires, which is very contrary to the way video games are built now with a 'one size fits many' approach,'" said Mark Riedl, co-creator of the model and assistant professor in the Georgia Tech School of Interactive Computing.
The model employs a performance arc, using data crunching to determine an algorithm which chooses in-game events designed to line-up predicted player performance with specifications for target performance, the latter in this case being finishing the game.
That's different to how video games currently react, with many using a technique known as 'rubberbanding' to make adjustments to in-game activities as a reaction to player performance. For example, a player doing badly at a car racing game will find that their computer controller opponents will slow down, giving them a chance to catch up.
"This is very reactionary," said Riedl, "You have to wait for things to fall apart, and then the game tries to correct it in this ad-hoc way."
"Our approach could allow novices to progress slowly and prevent them from abandoning a challenge right away," he added, "For those good at certain skills, the game can be tuned to their particular talents to provide the right challenge at the right time."
The same model has also been used to develop virtual missions for the US military, allowing soldiers' skills to be tested, however experienced they are.
"We've also done some work with the US Army," said Riedl, "to generate virtual missions where we choose and tailor the types of things that have to happen in the mission so that we don't overwhelm the novices or that we can really challenge the experts."
